Tweet
PROLOGUE
It seems to me that there are a several primary playing styles:
* Controller: This is often a warmonger, or someone who has "figured out" the game. Examples that come to mind include Vel, Arrian, Aeson, etc. One way or t'other, a very well-managed core empire, and manipulation of the AI civs around you. I think the "MY Continent" players often fit in here. Might be a hangover from Civ2. This style is very focused on what the OTHER civs are doing. I sort of think of this as playing a "winning style."
* Manager: This is typically thought of as Builder. I make a distinction here, however, as, yes, the building of cities and the empire is the player's focus, but inherent in that is a bit of an isolationist stance... I don't want to fight wars unless I have too; I'll trade but it isn't my game focus; I'll do my own research, thank you very much, etc. Often, this type of player will wait until Tanks or Cav before going military, and will even then only do so when fighting from a position of relative or absolute strength.
I think those are both great, and I'm especially impressed that through the years the different versions of Civ have been so obviously rewarding for VERY different styles, including presenting a balance of such complex factors (war vs. culture vs. research, etc.).
Ahhh, but therein may lie the problem.
I want something more... I'm calling it "games out of balance." (nods to Vel) I want grand epics... great challenges... the sweetness of victory and the ignominy of defeat. I want an enemy that I can respect... and hate through the ages.
The problem is, I think, that the default world settings, which were purposely designed to be "balanced" so that the different kinds of players would find satisfaction, seem to create too much of an "equal opportunity" world, whether for the player or the AI civs.
Forget difficulty level... that's setting the "against the odds" factor, although it does play a part in what I'm envisioning.
So I figured I'd take a shot at figuring out an answer.
OBJECTIVES
Think Greece and Persia. Rome and Carthage. China and Japan. Germany and France. England and France (the French are well, French). The US and Russia.
Essentially, I want a world that results in one or two killer AI civs, and me of course, to really duke it out over time.
But further: I want a world that looks like real life, over time, with winners and losers. A spectrum, if you will, of success, from schmuck to killer civs.
What I'm looking for is a way to set up benefits and detriments for AI civs right from the beginning of the game that will SELF-REINFORCE over time... getting the AI civs out of balance will hopefully result in the emergence of one or more killers.
[Note: This should also deal with two ancillary problems, tech trading and AI spending... the killer civs at the upper end of the spectrum pull away from the losers, and thus trading is reduced, with a further benefit that now the killers have money. Makes a big difference come the showdown].
APPROACH
Clearly, this is all about world settings.
What are we trying to achieve? Well, in summary:
* Uneven land quality: By this I mean some civs should have crap starts, and some great. But, wait, it gets better: Specifically, I want more hills and jungles... developing killers will have the ability to take advantage of these land types, and schmucks won;t.
* Uneven distribution of resources: Same as above, but worse, as the killers will take what they need, again and again.
* Large landmasses: It seems that on standard settings, especially for warmongers, it's too easy to either damage or destory all the civs around you. Like RL, I want continents that are big: big enough to support large civs, and big enough that those large civs are too far away for me to meaningfully impact early on.
* But not too much land: I DON'T want to play builder for 3000 years. Nor do I want the AIs doing that. The civs should be close enough for conflict, but not too close. Builders and Managers may not like it, but there needs to be early warfare to trip the aggression flag, otherwise the killers won't become killers (we've all seen it... you find the other continent in 500AD, and the 3 civs there have just about split it equally, and live in peace and harmony... I want one of'em to be a slavering pitbull).
THE MAP
Let's start with the map. I created 33 maps in various permutations using the editor, and then did various analyses of what I learned (I can't attach a spreadsheet; send me a PM if you want a copy). I know it's not a statistically valid sample, but it was good enough for my objectives.
So, first, starting conditions.
What I wanted was find out what settings would do the following:
Increase: Grassland, hills, mountains, jungles
Decrease: Plains, desert, tundra
[FYI, flood plains and forests don;t seem to vary much, although I suspect that there is a non-consistent increasen in the amount of flood plains when using default settings]
The tested maps were across the following categories:
Size Standard, Large, Huge
Water 70%, 60%
Climate Normal, Wet, Arid
Temperature Temperate, Cool, Warm
Age 4B, 3B
Using MapStat, here is what I selected as sort of the base case:
Size Standard
Water 70%
Climate Normal
Temperature Temp
Age 4B
% of Land Area (not including Coast)
Desert 9%
Plains 14%
Grassland 24%
Tundra 4%
Flood plains 3%
Hills 17%
Mountains 11%
Forest 14%
Jungle 6%
The combination that struck me as most closely matching the desired start state, disregarding map size and land coverage (I'll get back to those), was:
Climate Warm
Temperature Wet
Age 3B
% of Land Area (not including Coast)
Desert 2%
Plains 15%
Grassland 21%
Tundra 3%
Flood plains 1%
Hills 20%
Mountains 15%
Forest 13%
Jungle 10%
A note on age: 3 Billion works for a couple of reasons. First, it adds dramatically to the percentage of Hills tiles, which makes getting certain strategic resources more difficult (killers get it and schmucks don;t), and also they are much more valuable after Despotism, which again favors developing killers. More importantly, it clusters luxury resources... killers are more greatly incented to TAKE an area of land. The only real downside seems to be an overall slight decrease in Flood Plains.
Next, the size and configuration of the map, together with the number of players.
Although there are some terrain considerations when it comes to size (the bigger maps get, the more jungle as a percentage of total land), the primary considerations are
* distribution of resources
* distance between civs
[Note: I don't know why, but the Total Area reported by MapStat is not consistent with my understanding of map size. They are:
Standard 5,000 tiles
Large 8,450 tiles
Huge 12,800 tiles]
If we think of Standard, 70% water coverage, and 8 total civs, as the baseline, you get the following:
Total Land: 1,300 tiles (26% plus Coast tiles)
# Resource Tiles: 128
Resource / Total: 9.8%
Land per Civ: 163 tiles
In addition to the benefit of 3B years in clustering and difficulty of access, I wanted to decrease the percentage of resource tiles to total land, but without increasing the land per civ, and thus the average distance between them, too much.
Here are the settings I would pick, and results, depending on map size:
Standard
60% water
8 civs
7.0% resource tiles
228 tiles per civ
Large
70% water
10 civs
7.3% resource tiles
220 tiles per civ
Huge
70% water
15 civs
7.2% resource tiles
222 tiles per civ
I think the slightly greater distance between civs gives time for the inherent pluses or minuses of the various starting conditions to play out before military conflict, but does not result in a distance too great for early contact.
RESULTS AND IMPLICATIONS
For both the player and the AI civs, this approach to set-up obviously points to several civ selections... anything Industrious for example. In general, I would lean towards the most flexible civs, as you, the player, are subject to the spectrum of starting conditions as well.
If I'm right, this approach should work, resulting in killer and schmuck civs. Prior to sitting down and actually thinking it through, I had set up a number of games in this direction, and I've developed some unbelievable situations (the Egyptian Mess being probably the best example).
Again, by creating self-reinforcing positive conditions for CERTAIN civs, and poor conditions for others, basically unbalancing the whole game, the natural intercourse between civs should result in a spectrum of success. And again, if that happens, the killers get richer and stronger as they pull away, creating a greater challenge for the human player, when they inevitably meet.
And that's all without touching the editor.
So, in summary, if you're looking for a little more complexity, depth, strategy... or if you're just up for a challenge... try being a little unbalanced.
It seems to me that there are a several primary playing styles:
* Controller: This is often a warmonger, or someone who has "figured out" the game. Examples that come to mind include Vel, Arrian, Aeson, etc. One way or t'other, a very well-managed core empire, and manipulation of the AI civs around you. I think the "MY Continent" players often fit in here. Might be a hangover from Civ2. This style is very focused on what the OTHER civs are doing. I sort of think of this as playing a "winning style."
* Manager: This is typically thought of as Builder. I make a distinction here, however, as, yes, the building of cities and the empire is the player's focus, but inherent in that is a bit of an isolationist stance... I don't want to fight wars unless I have too; I'll trade but it isn't my game focus; I'll do my own research, thank you very much, etc. Often, this type of player will wait until Tanks or Cav before going military, and will even then only do so when fighting from a position of relative or absolute strength.
I think those are both great, and I'm especially impressed that through the years the different versions of Civ have been so obviously rewarding for VERY different styles, including presenting a balance of such complex factors (war vs. culture vs. research, etc.).
Ahhh, but therein may lie the problem.
I want something more... I'm calling it "games out of balance." (nods to Vel) I want grand epics... great challenges... the sweetness of victory and the ignominy of defeat. I want an enemy that I can respect... and hate through the ages.
The problem is, I think, that the default world settings, which were purposely designed to be "balanced" so that the different kinds of players would find satisfaction, seem to create too much of an "equal opportunity" world, whether for the player or the AI civs.
Forget difficulty level... that's setting the "against the odds" factor, although it does play a part in what I'm envisioning.
So I figured I'd take a shot at figuring out an answer.
OBJECTIVES
Think Greece and Persia. Rome and Carthage. China and Japan. Germany and France. England and France (the French are well, French). The US and Russia.
Essentially, I want a world that results in one or two killer AI civs, and me of course, to really duke it out over time.
But further: I want a world that looks like real life, over time, with winners and losers. A spectrum, if you will, of success, from schmuck to killer civs.
What I'm looking for is a way to set up benefits and detriments for AI civs right from the beginning of the game that will SELF-REINFORCE over time... getting the AI civs out of balance will hopefully result in the emergence of one or more killers.
[Note: This should also deal with two ancillary problems, tech trading and AI spending... the killer civs at the upper end of the spectrum pull away from the losers, and thus trading is reduced, with a further benefit that now the killers have money. Makes a big difference come the showdown].
APPROACH
Clearly, this is all about world settings.
What are we trying to achieve? Well, in summary:
* Uneven land quality: By this I mean some civs should have crap starts, and some great. But, wait, it gets better: Specifically, I want more hills and jungles... developing killers will have the ability to take advantage of these land types, and schmucks won;t.
* Uneven distribution of resources: Same as above, but worse, as the killers will take what they need, again and again.
* Large landmasses: It seems that on standard settings, especially for warmongers, it's too easy to either damage or destory all the civs around you. Like RL, I want continents that are big: big enough to support large civs, and big enough that those large civs are too far away for me to meaningfully impact early on.
* But not too much land: I DON'T want to play builder for 3000 years. Nor do I want the AIs doing that. The civs should be close enough for conflict, but not too close. Builders and Managers may not like it, but there needs to be early warfare to trip the aggression flag, otherwise the killers won't become killers (we've all seen it... you find the other continent in 500AD, and the 3 civs there have just about split it equally, and live in peace and harmony... I want one of'em to be a slavering pitbull).
THE MAP
Let's start with the map. I created 33 maps in various permutations using the editor, and then did various analyses of what I learned (I can't attach a spreadsheet; send me a PM if you want a copy). I know it's not a statistically valid sample, but it was good enough for my objectives.
So, first, starting conditions.
What I wanted was find out what settings would do the following:
Increase: Grassland, hills, mountains, jungles
Decrease: Plains, desert, tundra
[FYI, flood plains and forests don;t seem to vary much, although I suspect that there is a non-consistent increasen in the amount of flood plains when using default settings]
The tested maps were across the following categories:
Size Standard, Large, Huge
Water 70%, 60%
Climate Normal, Wet, Arid
Temperature Temperate, Cool, Warm
Age 4B, 3B
Using MapStat, here is what I selected as sort of the base case:
Size Standard
Water 70%
Climate Normal
Temperature Temp
Age 4B
% of Land Area (not including Coast)
Desert 9%
Plains 14%
Grassland 24%
Tundra 4%
Flood plains 3%
Hills 17%
Mountains 11%
Forest 14%
Jungle 6%
The combination that struck me as most closely matching the desired start state, disregarding map size and land coverage (I'll get back to those), was:
Climate Warm
Temperature Wet
Age 3B
% of Land Area (not including Coast)
Desert 2%
Plains 15%
Grassland 21%
Tundra 3%
Flood plains 1%
Hills 20%
Mountains 15%
Forest 13%
Jungle 10%
A note on age: 3 Billion works for a couple of reasons. First, it adds dramatically to the percentage of Hills tiles, which makes getting certain strategic resources more difficult (killers get it and schmucks don;t), and also they are much more valuable after Despotism, which again favors developing killers. More importantly, it clusters luxury resources... killers are more greatly incented to TAKE an area of land. The only real downside seems to be an overall slight decrease in Flood Plains.
Next, the size and configuration of the map, together with the number of players.
Although there are some terrain considerations when it comes to size (the bigger maps get, the more jungle as a percentage of total land), the primary considerations are
* distribution of resources
* distance between civs
[Note: I don't know why, but the Total Area reported by MapStat is not consistent with my understanding of map size. They are:
Standard 5,000 tiles
Large 8,450 tiles
Huge 12,800 tiles]
If we think of Standard, 70% water coverage, and 8 total civs, as the baseline, you get the following:
Total Land: 1,300 tiles (26% plus Coast tiles)
# Resource Tiles: 128
Resource / Total: 9.8%
Land per Civ: 163 tiles
In addition to the benefit of 3B years in clustering and difficulty of access, I wanted to decrease the percentage of resource tiles to total land, but without increasing the land per civ, and thus the average distance between them, too much.
Here are the settings I would pick, and results, depending on map size:
Standard
60% water
8 civs
7.0% resource tiles
228 tiles per civ
Large
70% water
10 civs
7.3% resource tiles
220 tiles per civ
Huge
70% water
15 civs
7.2% resource tiles
222 tiles per civ
I think the slightly greater distance between civs gives time for the inherent pluses or minuses of the various starting conditions to play out before military conflict, but does not result in a distance too great for early contact.
RESULTS AND IMPLICATIONS
For both the player and the AI civs, this approach to set-up obviously points to several civ selections... anything Industrious for example. In general, I would lean towards the most flexible civs, as you, the player, are subject to the spectrum of starting conditions as well.
If I'm right, this approach should work, resulting in killer and schmuck civs. Prior to sitting down and actually thinking it through, I had set up a number of games in this direction, and I've developed some unbelievable situations (the Egyptian Mess being probably the best example).
Again, by creating self-reinforcing positive conditions for CERTAIN civs, and poor conditions for others, basically unbalancing the whole game, the natural intercourse between civs should result in a spectrum of success. And again, if that happens, the killers get richer and stronger as they pull away, creating a greater challenge for the human player, when they inevitably meet.
And that's all without touching the editor.
So, in summary, if you're looking for a little more complexity, depth, strategy... or if you're just up for a challenge... try being a little unbalanced.
it would be great but then again , its "huge" , ....
Comment